NO20171396A1 - Cholesterol-producing yeast strains and their use - Google Patents

Description

The present invention relates to the production of cholesterol in organisms of the kingdom Fungi.

Cholesterol (see Figure 1) is the most important animal sterol. It is a fundamental component of cell membranes where it controls the cell membrane's fluidity and is present in all animal tissues and especially in nervous tissue.

Cholesterol is a product of considerable industrial interest. It is commonly used in the cosmetic industry. It is also used within the pharmaceutical industry, for example in drug delivery and also in cell culture.

Cholesterol is also used in the industrial synthesis of vitamin D3. This vitamin is then used to supplement human food (in dairy products, for example) and animal food. Cholesterol is also advantageously used as an additive in animal feed, especially in food intended for farmed shrimp.

At present, the majority of cholesterol sold is extracted from animal tissue (a small amount is produced by chemical synthesis). Two main sources are used for the extraction of cholesterol: spinal cord from cattle and lanolin, which is the natural fat in sheep's wool.

The use of animal tissue as a starting product causes problems. The recent problems associated with transmission of the prion responsible for scrapie to cattle (disease called BSE (bovine spongiform encephalitis) in cattle) have thus revived the need for caution when using animal tissue as a starting material. However, despite the steps taken, the risk of transmission of a pathogenic agent cannot be completely excluded. It would therefore be extremely beneficial to have a source of cholesterol that does not come from an animal tissue.

The purpose of the present invention is to provide a present source of cholesterol which is safe from a health point of view. The inventors have shown, surprisingly, that it is possible to redirect the natural production of ergosterol in Fungi to produce cholesterol.

General description of the invention

A first aspect of the invention relates to an organism from the kingdom Fungi which autonomously produces cholesterol.

A second aspect of the invention relates to an organism of the kingdom Fungi as described above, where the latter is genetically modified.

A third aspect of the invention relates to an organism of the kingdom Fungi as defined above. Where the latter produces cholesterol from a simple carbon source.

The invention also relates to an organism of the kingdom Fungi as defined above which expresses 7-dehydrocholesterol reductase and 3P-hydroxysterol 824 reductase enzymes. More specifically, the invention relates to an organism as defined above, where the sterol 24-C-methyltransferase enzyme has been inactivated and/or the C-22 sterol desaturase enzyme has been inactivated.

Another aspect of the invention relates to an organism from the kingdom Fungi as defined above, where the expression of 7-dehydrocholesterol reductase and the 3P-hydroxysterol 524 reductase enzymes is obtained by transforming the organism.

The invention also relates to an organism from the kingdom Fungi as defined above, where the inactivation of the sterol 24-C-methyltransferase enzyme is carried out by means of gene inactivation and/or the inactivation of the C-22 sterol desaturase enzyme is carried out by means of gene inactivation.

Another aspect of the invention relates to an organism from the animal kingdom Fungi as defined above, which is selected from the phylum Ascomycetes, more specifically from the subphylum Saccharomycotina, more specifically from the class Saccharomycetes or Schizosaccharomycetes, more specifically from the order Saccharomycetales or Schizosaccharomycetales, more specifically from the family Saccharomycetaceae or

Schizosaccharomycetaceae, even more specifically from the genus Sacharomyces or Schizosaccharomyces.

Another aspect of the invention relates to an organism from the kingdom Fungi as defined above, which is a yeast of the species Saccharomyces cerevisiae or Schizosaccharomycespombe.

The invention also relates to a method for producing cholesterol of non-animal origin, which comprises cultivating an organism as defined above. More specifically, in this method, the step of growing the organism is followed by a step of extracting the cholesterol. Preferably, the extraction of the cholesterol is carried out with a non-water miscible solvent.

More specifically, in the process as defined above, a saponification step is carried out prior to the extraction of the cholesterol. Even more specifically, in the method as defined above, a step consisting of mechanical crushing of the cells is carried out before the saponification or extraction of the cholesterol.

Another aspect of the invention relates to the use of an organism from the kingdom Fungi as defined above, to produce cholesterol or one of its metabolic intermediates or a mixture of sterols, labeled with 13C or with 14C.

The invention also relates to a method for producing the cholesterol or one of its metabolic intermediates, or a mixture of sterols labeled with<13>C or with<14>C, comprising the following steps: - growing an organism from the kingdom of Fungi as defined above on a <13>C-labeled or <14>C-labeled substrate, and - to extract said cholesterol or one of its metabolic intermediates or the mixture of sterols.

The invention also relates to a method for producing an isotopic mixture of cholesterol, of cholesterol intermediates or of cholesterol metabolites, labeled by different processes using isotopic labeling, which comprises growing an organism from the kingdom Fungi as defined above on a labeled substrate and then on an unlabeled substrate , the cultivation time on each of these substrates has been chosen to provide a defined isotopic profile. The invention also relates to a sample of molecules of cholesterol, of cholesterol intermediates or of cholesterol metabolites marked at different positions by using isotopic markings, which have a defined isotopic profile and which can be obtained by means of this method for production.

The invention also relates to a composition which contains, as a traceable label, an isotopic mixture of cholesterol, of cholesterol intermediates or of cholesterol metabolites, labeled at different positions by using isotopic labels and which has a defined isotopic profile. More specifically, this composition is intended for the field of human or animal food or therapy.

Detailed description of the invention

The present invention relates to the production of cholesterol in organisms of the kingdom Fungi. In Fungi, no cholesterol is found in the natural environment. The last mentioned is an animal sterol. The main sterol of the cell membranes of these organisms is ergosterol.

The present invention makes it possible to carry out cholesterol synthesis, through the multiplication of Fungi, in the presence of a simple carbon source. The method proposed by the present invention therefore makes it possible to provide a large amount of cholesterol, at low costs, because the method uses the cultivation of organisms from the kingdom Fungi and the addition of a simple carbon source, which is readily available commercially.

According to the present invention, the expression "simple carbon source" is intended to mean carbon sources which can be used by those skilled in the art for the normal growth of a fungus and especially of a yeast. It is intended to denote in particular the various sugars, such as glucose, galactose or sucrose or molasses, or the by-products of these sugars. A simple carbon source that is most preferred is ethanol and glycerol.

The fact that the production is carried out autonomously means that there is no need to add substrates to provide the cholesterol, but that the organism can produce cholesterol only from the initially simple carbon source. It is also clear that the strain can produce the cholesterol using a substrate located upstream in the metabolic pathway, as long as the strain of the organism according to the present invention contains all the genes necessary to complete the metabolic pathway for cholesterol production.

The invention relates in particular to a genetically modified organism from the kingdom Fungi (a Fungus) which autonomously produces cholesterol from a simple carbon source.

A certain number of genetic modifications in the fungus can be effected to divert the natural metabolic reaction pathway of ergosterol production towards the production of cholesterol. The present invention thus relates to a genetically modified organism from the kingdom of Fungi which expresses the 7-dehydroxycholesterol reductase and 3P-hydroxysterol 824 reductase enzymes. The organism strain from the kingdom Fungi thus produces modified cholesterol. The applicant has in fact been able to modulate, by means of the results obtained (cf. the example section of the present application) the metabolic reaction pathway resulting in ergosterol, and in some of its derivatives (cf. Figure 2). Expression of the 7-dehydrocholesterol reductase and 3P-hydroxysterol 824 reductase enzymes in the fungus S. cerevisiae may allow the production of cholesterol by diverting part of its biosynthetic pathway and ergosterol.

The 7-dehydrocholesterol reductase enzyme carries the number EC: 1.3.1.21 in the International Enzyme Classification. It is also called 5-5,7-sterol-8-7-reductase, 7-DHC-reductase or Sterol 8-7-reductase and will also be called 87-sterol reductase, 8-7red, 8-7-reductase or 87 -reductase in the remainder of this document. This enzyme catalyzes, in the natural state in plants, for example the NADPH-dependent reduction of delta-5,7-cholestadienol to delta-5-cholestaneol or the reduction of sterol intermediates that have the double bond in the 7-8 position (Taton and Rahier, 1991). The gene encoding the 7-dehydrocholesterol reductase enzyme was isolated for the first time in the plant Arabidopsis thaliana; the isolation of the corresponding gene and the expression of this enzyme in the yeast Saccharomyces cerevisiae is described in patent EP 727 489. The sequences of this gene and of the protein are available under the following gene bank accession number: U49398 (Lecain et al., 1996).

A certain number of homologues of this gene have been described in other species. These are, for example, the homologous gene in humans (nucleotide sequences of which are available under Genbank number AF034544, the protein sequence is available under Genbank number: AAC05086) (Moebius et al., 1998); the homologous gene of the rat Rattus norvegicus (the nucleotide sequence of the gene is available under Genbank number: AB016800, the protein sequence of which is available under Genbank number: BAA34306). Homologous genes have also been identified in the chicken Gallus gallus, with gene bank reference BM490402 or in the toad Xenopus laevis, with gene bank reference BI315007, or the zebrafish Danio rerio, with gene bank reference BQ 132664. A gene encoding a delta7 sterol reductase activity is also found in plants such as rice , Oryza sativa, with genebank reference CA753545, or potato, Solarium tuberosum, with genebank reference BF342071. This gene encoding a delta7 sterol reductase activity can also be found in the protist Mastigamoeba balamuthi, with the gene bank reference BE636562.

Those skilled in the art will be able to readily isolate other homologous genes encoding the 7-dehydrocholesterol reductase enzyme in other organisms. They may in particular refer to the cloning method described in Example 1 of patent EP 727 489, which describes a cloning method for isolating a cDNA encoding a protein having delta-5,7-sterol-delta-7-reductase activity. Those skilled in the art can also easily determine the 7-dehydrocholesterol reductase activity in the corresponding proteins, in particular by applying the activity assay also described in example 1 of patent EP 727 489.

Expression of the 7-dehydrocholesterol reductase enzyme in an organism from the kingdom of Fungi according to the invention can be obtained by any means known to those skilled in the art. These may in particular involve transforming the organism with a construct comprising an expression cassette occupied by a transcription promoter, preferably homologous, from the open reading frame encoding the 7-dehydrocholesterol reductase enzyme and by an appropriate transcription terminator, according to the usual rules known to those skilled in the art . As a homologous promoter, use will generally be made of a promoter suitable to allow sufficient and functional expression of the heterologous protein. The promoter can be, for example, the PGK promoter, the ADH promoter, the CYC1 promoter, the GALI0/CYCl promoter, the TDH3 promoter or the TPI promoter. The terminator can, for example, be the terminator of the phosphoglycerate kinase (PGK) gene. Said expression cassette may be integrated, in the form of one or more copies, into the nuclear or mitochondrial genome of the host, or it may be carried by an artificial structure of the yeast artificial chromosome (YAC) type or carried by an episomal genetic element such as a plasmid. To effect this type of expression, yeast of the Yarrowia lipolitica, Kluyveromyces lactis or Pichia pastoris type can be used, for example.

Preferably, the 7-dehydrocholesterol reductase enzyme that is expressed is the enzyme from the plant Arabidopsis thaliana (an example of a method for expressing this enzyme in the yeast Saccharomyces cerevisiae is described in patent EP 727 489). However, it can be any homologous or non-homologous, natural or artificial, enzyme that exhibits the same enzyme activity.

The 3P-hydroxysterol 824 reductase enzyme, also called DHCR24 or 24-dehydrocholesterol reductase, naturally catalyzes the reduction of desmosterol (cholesta-5, 24-dienol) or to lanosterol derivatives that have a double bond in the 24-25 position of the side chain (for example, 14-desmethyl- lanosterol, zymosterol or cholesta-7, 24-dienol), the reduction of which is necessary for the biosynthesis of cholesterol in humans in particular (HR. Waterham et al., 2001). This enzyme will also be called delta 24-(25) sterol reductase, delta 24 sterol reductase or 824 reductase in the remainder of this document.

The gene encoding the 3P-hydroxysterol 524 reductase enzyme was isolated for the first time in humans; the isolation of the corresponding gene and the expression of this enzyme in the yeast Saccharomyces cerevisia is described in the publication HR. Waterham et al., 2001. The frequencies of this gene, and of the protein, are available under the following Genbank accession numbers: NM_014762 and NP_055577.

A certain number of homologues of this gene have been described in other species. They are, for example, the homologous gene in mice (Mus musculus) (the nucleotide sequence of which is available under Genbank number: NM 053272, the protein sequence of which is available under Genbank number: NP 444502. Homologs have been described in the worm Caenorhabditis elegans, and in particular a complementary DNA with the genebank reference AF026214. Homologous sequences have also been described in plants, such as cotton, Gossypium hirsutum, with the genebank reference AAM 47602.1, rice, Orysa sativa, with the genebank reference AAP53615, or pea, Pisum sativum, with the genebank reference A AK15493.

Those skilled in the art will be able to readily isolate other homologous genes encoding the 3P-hydroxysterol 824 reductase enzyme in other organisms. You can refer in particular to the cloning procedure described in the publication HR. Waterham et al., 2001. Those skilled in the art will also be able to readily determine the 3P-hydroxysterol 824 reductase activity of the corresponding proteins, particularly by applying the activity assay also described in the publication (Waterham et al., 2001).

Expression of the 3 P-hydroxysterol 824 reductase enzyme in an organism of the kingdom Fungi according to the invention can be obtained by any means known to those skilled in the art. This may particularly involve the means described above with respect to the expression of the 7-dehydrocholesterol reductase enzyme.

Preferably, the 3 β-hydroxy sterol 824 reductase enzyme expressed is the human enzyme. An example of isolation of the corresponding gene, and of the expression of this enzyme in the yeast Saccharomyces cerevisiae, is described in the publication HR. Waterham et al., 2001. However, it can be any homologous or non-homologous, natural or artificial, enzyme that exhibits the same enzyme activity.

The organisms of the kingdom Fungi, according to the present invention, advantageously express the 7-dehydrocholesterol reductase and 3P-hydroxysterol 824 reductase enzymes and also exhibit inactivation of the sterol 24-C-methyltransferase enzyme.

The sterol 24-C-methyltransferase enzyme carries the number EC-2.1.1.41 in the International Classification of Enzymes. It is also called ERG6p, delta(24)-methyltransferase, delta(24)-sterol methyltransferase, zymosterol-24-methyltransferase, S-adenosyl-4-methionine:sterol delta(24)-methyltransferase, SMT1, 24-sterol C-methyltransferase , S-adenosyl-L-methionine:delta(24(23))-sterol methyltransferase or phytosterol methyltransferase. This enzyme naturally catalyzes the C-24 methylation of zymosterol, resulting in the formation of fecosterol.

The gene that codes for the sterol 24-C-methyltransferase enzyme was named Erg6 in the yeast Saccharomyces cerevisiae. The sequence of this gene is available under the following Genbank accession number: NC 001145. The sequence of the corresponding protein is available under the following Genbank accession number: NP_013706 (Bowman et al., 1997), (Goffeau et al., 1996).

A certain number of homologues of this gene have been described in other Fungi. They are, for example, the homologous gene in Schizosaccharomycespombe (the nucleotide sequence of which is available under Genbank number Z99759, the protein sequence of which is available under Genbank number: CAB16897 (Wood et al., 2002); the homologous gene in Neurospora crassa (the nucleotide sequence of which is available under Genbank number: NCB24P7, the protein sequence of which is available under Genbank number: CAB97289); the homologous sequence in Candida albicans (the nucleotide sequence of which is available under Genbank no. AF031941, the protein sequence is available under Genbank number: AAC26626) (Jensen-Pergakes et al., 1998) Genes encoding an enzyme homologous to ERG6 have also been described in Candida lusitaniae, at gene bank reference AA021936.1 and also in Pneumocystis carinii (Kaneshiro et al., 2002) or in Kluveromyces lactis (Ozier-Kalogeropoulos et al., 1998).

Those skilled in the art will readily be able to isolate other genes homologous to the Erg6 gene in organisms from the kingdom Fungi. Those skilled in the art will also be able to easily determine the sterol 24-C-methyltransferase activity of the corresponding proteins, in particular by using, as an activity assay, the functional complementation of a yeast strain disrupted in these genes. The complementation is then reversed by the formation of sterols which are branched at the 24-position, particularly sterols of the ergosta type bearing a methylene group at the 24-28 position. The presence of ERG6-type sterol 24-C-methyltransferase biological activity will also be determined in vitro using the techniques developed by (McCammon et al., 1984) or by Taylor and Parks (Taylor and Parks, 1978). Furthermore, the sterols that are produced, and the substrate for the ERG6 enzyme, will be separated by gas chromatography according to the technique developed by Nes in (Methods in Enzymology Steroids and Isoprenoids Volume 111 part B, 1985, "A comparison of Methods for the Identification of Sterols" , pp. 3-37).

The organism strain of the kingdom Fungi according to the present invention which expresses the cholesterol reductase and 3P-hydroxysterol 824 reductase enzymes, and which also inhibits inactivation of the sterol 24-C-methyltransverase enzyme produces cholesterol. Applicant has indeed been able to determine that, surprisingly, the inactivation of the sterol 24-C-methyltransferase enzyme blocks the upstream ergosterol biosynthetic pathway, allowing increased production of cholesterol by the fungus strain (compare the examples section of the present application).

The 7-dehydrocholesterol reductase and 3 β-hydroxy sterol 824 reductase enzymes are expressed as described above.

The inactivation of the sterol 24-C-methyltransferase enzyme can be carried out by means known to those skilled in the art. It may in particular involve the introduction, by means of mutagenesis, of a nonsense mutation, of an insertion or of a setting which caused a change in the reading frame of the gene encoding said protein.

It may also involve the expression of an antisense RNA complementary to the messenger RNA encoding said protein or the gene silencing system known to those skilled in the art as RNAi (small interfering RNA) and the associated enzyme systems if these are not naturally present in the host. The mutagenesis can be carried out in the coding sequence or in a non-coding sequence to render the encoded protein inactive or to prevent its expression or its translation. The mutagenesis can be carried out in vitro or in situ by suppression, substitution, deletion and/or addition of one or more bases in the gene in question, or by means of gene inactivation.

In particular, this may involve the introduction of an exogenous DNA into the coding or promoter sequence (eg an expression cassette with a homologous promoter and/or terminator and a heterologous coding part). The expression cassette advantageously allows expression of a selection marker. It is also possible to modify the promoter of the gene to reduce the expression level. For fungi, inactivation is also performed by interrupting the coding sequence with the coding sequence of a heterologous or homologous marker gene. The main techniques for disrupting a gene from fungi are described in the article by Johnston et al., (2002) (Methods in Enzymology, volume 350, published by Christine Guthrie and Gerry Fink; "Gene Disruption"; M. Johnston, L. Riles, J. Hegemann, pp. 290-315).

The organisms in the kingdom Fungi according to the present invention advantageously express the 7-dehydrocholesterol reductase and 3P-hydroxysterol 824 reductase enzymes and also inhibit inactivation of the C-22 sterol desaturase enzyme. The C-22-sterol desaturase enzyme is also called ERG5p, Cyp61, cytochrome p-45061 or sterol delta22-desaturase. This enzyme naturally catalyzes the conversion of ergosta-5, 7, 24 (28)-trienol to ergosta-5, 7, 22, 24 (28)-tetraenol by adding a double bond at position C22 (cf. Figure 2).

The gene that codes for the C-22 sterol desaturase enzyme was named Erg5 in the yeast Saccharomyces cerevisiae. The sequence of this gene is available under the following Genbank accession number: U34636. The sequence of the corresponding protein is available under the following Genbank accession numbers: AAB06217 (Skaggs et al., 1996) or P54781 (Bowman et al., 1997).

A certain number of homologues of this gene have been described in other Fungi. They are, for example, the homologous gene in Schizosaccharomycespombe (the nucleotide sequence of which is available under Genbank number Z98974, the protein sequence of which is available under Genbank number CABl 1640) (Wood et al., 2002); the homologous gene in Symbiotaphrina buchneri (the nucleotide sequence of which is available under Genbank number: AB086896, the protein sequence of which is available under Genbank number: BAC01142)

(Noda and Koizumi, 2003); the homologous gene in Symbiotaphrina kochii (the nucleotide sequence of which is available under Genbank number: AB086890, the protein sequence of which is available under Genbank number: BAC01139) (Noda and Koizumi, 2003); the homologous gene in Candida albicans

(the nucleotide sequence of this is available under Genbank number: AL033396, the protein sequence of this is available under Genbank number: CAA21953) (Tait et al., 1997). The ERG5 gene has also been described in Candida lusitaniae, with genebank reference AAO48601.

Those skilled in the art will readily be able to isolate other genes homologous to the £>*,g5 gene in organisms of the kingdom Fungi. Those skilled in the art will also readily be able to determine the C-22 sterol desaturase activity of the corresponding proteins, particularly using the activity assay described by B. A. Skaggs et al., 1996. This activity can also be demonstrated by functional complementation of a S. cerevisiae - yeast that is destroyed in advance in the erg5 gene. This complementation will be attested by the presence, in the complemented strain, of ergosta-5, 7, 22-trienol. The C22-sterol desaturase activity can be measured in vitro using the method described by Kelly and Baldwin et al., JBC (1997), after lysis of the yeast (Kelly et al., 1997). ;The organism strain of the kingdom Fungi according to the present invention which expresses the 7-dehydrocholesterol reductase and 3P-hydroxysterol 824 reductase enzymes and which also inhibits inactivation of the C-22 sterol desaturase enzyme produces cholesterol. Indeed, the applicant has been able to determine that the inactivation of the C-22-sterol desaturase enzyme advantageously blocks the conversion of cholesterol to cholesta-5,22-dienol and allows the stabilization of the production of cholesterol (cf. the examples section of the present invention). This blockage also takes place at the level of the conversion of cholesta-5,7-dienol, a precursor of cholesterol, to cholesta-5,7,22-trienol, a precursor of cholesta-5,22-dienol. Surprisingly, the C-22-sterol desaturase enzyme actually accepts cholesterol as a substrate and converts it to cholesta-5, 22-dienol. This parasitic reaction can be eliminated by inactivating the C-22-sterol desaturase enzyme, which the applicant has been able to determine. ;The expression of the 7-dehydrocholesterol reductase and 3P-hydroxysterol 824 reductase enzymes is carried out as described above. The inactivation of the C-22 sterol desaturase enzyme can be accomplished by any means known to those skilled in the art. They may in particular be the methods described above with regard to the inactivation of the sterol 24-C-methyltransferase enzyme. ;The organisms in the kingdom Fungi according to the present invention advantageously express the 7-dehydrocholesterol reductase and 3P-hydroxysterol 824 reductase enzymes and also exhibit inactivation of the C-22-sterol desaturase enzyme and inactivation of the sterol 24-C-methyltransferase enzyme. These strains actually inhibit the cumulative benefits of the two inactivations and are cholesterol-producing strains. ;The expression of the 7-dehydrocholesterol reductase and the 3P-hydroxysterol 824 reductase enzymes and the inactivation of the C-22-sterol desaturase and the sterol 24-C-methyltransferase enzymes are carried out as described above. ;In one embodiment, the cholesterol is present in the organism strain according to the present invention in a part greater than 20%, preferably 35%, most desirably 50% or more of the total sterols produced by the strain according to the invention (especially the synthesis intermediates ). Preferably, organisms in the kingdom Fungi according to the present invention are selected from the phylum Ascomycetes, more desirably they are selected from the subphylum Saccharomycotina, even more desirably they are selected from the class Saccharomycetes or Schizosaccharomycetes, even more desirably they are selected from the order Saccharomycetales or Schizosaccharomycetales , even more desirably they are selected from the family Saccharomycetaceae or Schizosaccharomycetaceae, even more desirably they are selected from the genus Saccharomyces or Schizosaccharomyces, completely desirably former organism in the kingdom Fungi according to the invention the species Saccharomyces cerevisiae or Schizosaccharomyces pombe. ;The present invention also relates to a method for producing cholesterol from non-animal origin, which comprises the following steps: ;an organism from the kingdom Fungi as defined above was cultivated, ;- the cholesterol produced by this organism is extracted. ;The extraction is based on the treatment of the fungus with a solvent for cholesterol, preferably a non-water-miscible solvent. This treatment can ideally be combined with any method for mechanically crushing the cells. More desirably, the fungus will be treated, prior to extraction with the solvent, with a saponification mixture intended to release the cholesterol possibly bound to other cellular compounds such as fatty acids in particular. This saponification mixture may consist of a base, for example aqueous ammonia, sodium hydroxide or potassium hydroxide, dissolved in water or more desirably in a water-miscible organic solvent such as, for example, methanol or ethanol, or a solvent-water mixture. The saponification can be carried out without or preferably with heating to a temperature of 60-120°C, at atmospheric pressure or at low pressure. The extraction with the non-water soluble solvent can be replaced with a solid phase extraction on a hydrophobic resin. A sterol extraction procedure is described by L. Parks et al., (1985) (Methods in Enzymology 111 unpublished by L. Rilling, L. Parks, C. Bottema, R. Rodriguez and Thomas Lewis, pp. 333-339). ;The crude cholesterol thus obtained can be purified by any method known to those skilled in the art, especially that described by Boselli E, Velazco V, Caboni Mf and Lercker G J, Chromatogr A, May 11, 2001.; 917(l-2):239-44. Other methods can also be used, such as that described for the extraction of cholesterol from the wool of sheep. Those skilled in the art may refer in particular to the methods described in US Patents US 2,688,623 or US 2,650,929 or in British Patents GB690879, GB646227 or GB613778. Another aspect of the invention relates to the use of the strains according to the present invention to provide cholesterol or one of its metabolic intermediates, or a labeled mixture of sterols. The term "metabolic intermediate of cholesterol" is intended to mean in particular the sterols specified in Figure 2. They may in particular be cholesta-8,24(25)-dienol, cholesta-7,24(25)-dienol, cholesta-5, 7,24(25)-trienol, cholesta-5,24(25)-dienol or cholesta-5,22-dienol. The principle for providing a labeled cholesterol is described in figure 10. This manipulation consists in first of all growing the Jungus strain on a fully labeled substrate. The cells are then cultured for an unlabeled substrate. There is thus a change in the isotopic labeling of the carbon source; there follows the "ovosynthetic metabolic intermediates and then of sterol, including cholesterol, and includes a gradual change in labeling. This therefore involves a profile that is complex but can be fully determined experimentally and that represents a unique isotopic signature that is dependent simultaneously: 1) on the labeling protocol and in particular culture times and conditions with labeled and unlabeled substrate, ;2) on the exact genetic structure of the strain used, ;3) at the precise time at which the cultures are stopped. ;Once the culture has been stopped (for example, by cell lysis or by stopping the culture in the presence of a sublethal concentration of cytotoxic or cytostatic antifungal products), the labeled cholesterol or one of its metabolic intermediates, or a labeled mixture of sterols, is extracted and cleaned as described above. ;The isotopic profile of the labeled cholesterol, or of one of its metabolic intermediates, or of the labeled mixture of sterols, has several unique properties: 1) it can be modulated as desired by adjusting culture conditions, the strain used and the sterol selected. A unique marking register can therefore be produced; 2) it is "combinable", that is, multiple isotopic signatures corresponding to multiple unique sterols labeled with isotopic profiles that can themselves be modulated can be combined to form a "molecular alphabet"; 3) it is reproducible and easy to determine experimentally; 4) it corresponds to a molecular tracer compound that is easy to isolate, stable, colorless and odorless, non-volatile and non-toxic, and which can be incorporated into food, a medicinal product, additives or other products that can be assimilated by humans; 5) it cannot be falsified without having the specific recombinant strains and the very precise labeling, cultivation and extraction conditions. In addition, knowledge of the isotopic signature does not make it possible to trace back to the parameters that made it possible to produce it. A "isotope alphabet" for general use, which cannot be falsified and which can be incorporated into the products of any type, including consumer goods, can thus be readily provided by means of the present invention. There is, so to speak, an unlimited number of "isotopes" that can be assembled from such an alphabet by making use of both the labeling profiles and the different types of sterols. The incorporation of such signatures into the most varied products therefore contributes to a unique method of labeling that cannot be counterfeited. Unlike, for example, DNA signatures which can be reproduced as soon as they are known. The signature can also be read non-destructively, for example by laser ionisation followed by mass spectrometry analysis (MALDI-TOF or similar). ;The use of<13>C-labeled substrate instead of the unlabeled carbon sources to grow the./wwgw.s strains according to the invention makes it possible to synthesize very strongly labeled sterols, and especially cholesterol (comprising mine 95% of <13>C carbon). The preparation of<14>C radioactive sterols and cholesterol is also possible using the same approach. The method can also be incorporated into yeast strains that produce steroids, and especially hydrocortisone (cf. patent application WO 02/061109), to produce labeled or <14>C-labeled steroids, for example for RIA assays. ;Figure Captions ;Figure 1: Chemical formula of cholesterol, and also the nomenclature commonly used to number the different carbons and the hub of the different rings. The four rings of the cholesterol molecule are called A, B, C and D, respectively, and the carbons are numbered from 1 to 27. Figure 2: Simplified diagram of the last part of the biosynthetic reaction pathway for sterols of the ergosta and cholestatypes in natural or modified yeast. The scheme is not exhaustive, but allows defining the steps involving the enzymes mentioned in this document. The ERG2p, ERG3p, ERG5p and ERG6p proteins are fungus or yeast proteins while delta-7red (delta-7 sterol reductase) and delta 23 (25)red (delta 24-(25) sterol reductase) proteins are heterologous proteins of mammalian origin or of plant origin. Figure 3: Compared HPLC profile with UV detection at 206 nm, of the free sterols in the strains derived from the BMA64 strain and identification of these sterols. The strains studied are as follows: WGIF01 (BMA64 strain disrupted in the erg6 gene (cf. Example 1)), WGIF02 (BMA64 strain disrupted in the ergtf gene and expressing the 524 reductase, Example 12), WGIF03 (BMA64 strain disrupted in the erg6 gene and expressing 87-reductase, Example 13), WGIF04 (BMA64 strain disrupted in the ergtf gene and expressing the 57-reductase and 824-reductase, Example 14). C5: cholesta-5-enol (cholesterol); C5,22: cholesta-5,22-dienol; C5,24: cholesta-5,24-dienol (desmosterol); C8,24: cholesta-8,24-dienol (zymosterol); C5,7,22: cholesta-5,7,22-trienol; C5,7,24: cholesta-5,7,24-trienol; C5,22,24: cholesta-5,22,24-trienol; C5,7,22,24: cholesta-5,7,22,24-tetraenol; lan: lanosterol. Figure 4: Comparative HPLC profile with UV detection at 206 nm, of the free sterols of the WGIF04 strain (BMA64 strain disrupted in the ergtf gene and expressing the 87-reductase and the 824-reductase, example 14) after 0.2 , 4, 8 and 24 h induction with galactose. 8: WGIF01 strain (Example 1). For the WGIF04 strain, the samples are taken at 0, 2, 4, 8 and 24 hours after switching the carbon source to galactose. The profile of the BMA64 strain carrying the ergtf disruption (WGIF01) presented is that obtained immediately after the switch to galactose. This profile remains virtually unchanged during induction (0-24 hours). The absorption signal at 206 nm corresponds to the absorption coefficients which are variable from one sterol to another. C5: cholesta-5-enol (cholesterol); C5,22: cholesta-5,22-dienol; C5,24: cholesta-5,24-dienol (desmosterol); C8,24: cholesta-8,24-dienol (zymosterol); C5,7,22: cholesta-5,7,22-trienol; C5,7,24: cholesta-5,7,24-trienol; C5,22,24: cholesta-5,22,24-trienol; C5,7,22,24: cholesta-5,7,22,24-tetraenol; lan: lanosterol. Figure 5: Comparative HPLC profile with positive ionization electron spray detection (mass spectrometer), for the free sterols of the WGIF04 strain (Example 14) after 0, 2, 4, 8 and 24 hours of induction with galactose. 8: WGIF01 strain. C5: cholesta-5-enol (cholesterol); C5,22: cholesta-5,22-dienol; C5,24: cholesta-5,24-dienol (desmosterol); C8,24: cholesta-8,24-dienol (zymosterol). The HPLC profiles come from the same analyzes as those in Figure 4. ;Figure 5A (left): Detection at m/z = 367, Figure 5B (right): m/z = 369. ;Y-axis: number of ions counted per second. X-axis: elution time in minutes. ;Figure 6: Details of the profile at m/z = 369 by HPLC for the three strains. WGIFOl, CA10 carrying the expression plasmid of delta 24 sterol reductase and for WGIF04, cholesterol is injected as an internal standard. The amounts of total sterols injected for the three strains correspond to extractions performed on identical amounts of culture as measured by absorbance at 600 nm. Figure 7: Comparative profiles of the total sterols (free and esters), by gas chromatography, of WGIFOl (deletion of erg6), WGIF02 (deletion of erg6 with expression of the 824 reductase), WGIF03 (deletion of erg6 with expression of 87 -reductase), WGIF04 (deletion of erg6 with expression of the 824-reductase and 87-reductase) and CA10 pYES_Delta24 (FYl 678 genetic background, deletion of erg5 with expression of 824-reductase, 87-reductase, erg5) strains. The response scales (flame ionization currents) are arbitrary. The profiles should only be compared qualitatively from one strain to the other. However, the retention time scale is the same for all strains (retention times are expressed in minutes). The sterols have been identified according to the criteria described in the present application. Figure 8: Quantitative distribution of the main free sterols in the yeast strains (BMA64 (Figure 8A), WGIFOl (Figure 8B), WGIF02 (Figure 8C) and WGIF03 (Figure 8D) evaluated on the basis of the UV spectra. The distribution is given as a percentage of the total species presented in the figure, and which are the only ones that can be detected in appreciable amounts. In the absence of a standard for several of the intermediate sterols, the quantification remains performed on the basis of UV spectra associated with each of the peaks of the HPLC chromatogram by applying the evaluated absorption coefficients given below (cf. Table 1, the absorption coefficients expressed in mM per liter and per cm). To do this, the absorption coefficients, which correspond to the unsaturated structural units present in the structure, of a given sterol are tried in Table 1 and optionally added (if multiple units are present in the same molecule) to provide an evaluation of the extinction coefficient of each type of sterol.The evaluation was done by applying the values at 280 nm if at least one unit absorbed at this wavelength is present, if this fails the wavelength 235 nm is used, and if absorption at the last mentioned wavelength fails, then the wavelength 206 nm is used to evaluate the concentrations of each of the sterols from the respective absorption signals by HPLC. Figure 9: Quantitative distribution of the main free sterols in the WGIF4 yeast strain, evaluated on the basis of UV spectra. The quantifications are carried out in the same way as described in figure 8. ;Figure 10: Principle for isotopic labeling of sterols by substitution of carbon sources. ;Figure 11: Evaluation of the isotope labeling profiles of the cholesterol produced in the WGIF04 strain after 4, 8 and 24 hours of induction. The free sterols are extracted and separated by HPLC as described. A mass spectrum between the values of m/z = 300 and m/z = 450 is obtained every 0.2 seconds during the elution. These spectra were then averaged over windows of 1.8 seconds, and then subjected to a multilinear regression using, as the regression base, a set of 24 vectors representing the theoretical mass distributions of the labeled cholesterol for a random incorporation, by independent selection, of carbon 13 at each of the 27 processions in the molecule with an option to label each carbon variable between 0 and 1 according to the vector considered. The marking probabilities for the different vectors used as a base are chosen so that the cross-correlation coefficient for the distributions of two subsequent vectors to the base is 0.92, the base starts with a vector corresponding to a probability of presence of 100% at all the processions on carbon 12. The multilinear adjustment is done on the basis of a least square statistical criterion, earning the non-diagonal expressions of the matrix will be the products of the partial derivatives of the Gauss progression (maximum numerical filtering). After analysis, the mass spectrum is then reconstructed on the optimized basis. The curves represented in the figures therefore represent the result of the optimally filtered reconstruction after standardizing the maximum amplitude to the value 100. ;For each induction time point, the two curves represent two independent profiles corresponding to elution times that differ by 1.8 seconds and which correspond to spectra localized in the central zone of the cholesterol elution peak. The figure shows that the analysis is very reproducible. Figure 12: Example of different isotope signatures with different sterols or different induction times. The same calculation and representation as in Figure 11, but for different sterols and different induction times. The value of RT indicates the range of retention time used for the calculation (in minutes). The values for this range are as follows: Fig. 12 A: RT = 12.25-12.42, Fig. 12 B: RT = 12.2-12.7, Fig. 12 C: RT = 12.25-12.35, Fig. 12 D: RT = 13.3- 13.6. The induction times are 8 or 24 hours. ;The values of m/z indicate the left and right limits of the m/z values. The lowest value from the m/z for each box corresponds to the m/z of the sterol made only from carbon 12 Jan. Figure 13: Comparative profiles of the total sterols (free and esters), in gas chromatography, of the YIM59/pIM303 strain (part A of the figure) and of the YIM59/pIM331 strain (part B of the figure) (Example 18). The response scales are arbitrary. The retention time scale is the same from both strains (retention time is expressed in minutes). The sterols are identified according to the criteria described in the present application. The present invention is illustrated by using the following examples, which must be regarded as non-limiting illustrations. ;The molecular, biological techniques used are described by Ausubel et al., some yeast manipulations are described by Adams et al., (Adams and Holm, 1996). Example 1: Construction of a S. serevisiae yeast strain with a disruption in the erg6 gene (WGIF01 strain): The S. cerevw/ae yeast strain WGIFOl in which the ERG6 gene is disrupted with the TRP1 gene was obtained by transforming The BM64 strain with a PCR product carrying a functional TRPl gene flanked by extremities homologous to the ERG6 gene; the BM64 strain (with genotype MA Ta; ura3- 52; trplA2; leu2- 3_ 112; his3- ll ; ade2- I; can 1- 100) is a derivative of the S. cerevisiae yeast strain W303 MATa by complete deletion of the TRP1 gene. The BMA64 strain and the W303 MATa strain are described in the publication by Baudin-Baillieu et al. (Baudin-Baillieu et al., 1997). To isolate the TRP1 gene, the TRP1 gene in plasmid pFL44 (Bonneaud et al., 1991) was transformed using Z-Taq1 (a DNA-dependent DNA polymerase) provided by Takara (PanVera LLC 501 Charmany Drive, Madison , WI 53719 USA). The pair of primers used makes it possible to transform, by means of the DNA polymerase, the TRP1 gene surrounded by sequences corresponding to ERG6 geneX. ;The sequence of these primers is as follows: ; ; The PCR (polymerase chain reaction) product thus obtained is purified by electroelution of the fragment corresponding to the expected size, and is used to transform the BM64 strain by means of the lithium chloride technique as described by (Gietz et al., 1995). ;After transformation, the treated yeasts are plated on minimal medium without tryptophan (Gietz et al, 1995). 41 transformed BM64 colonies which are prototrophic for tryptophan are thus provided. These 41 cholines were then tested for three of their properties: sensitivity to nystatin, genomic structure of the insertion of the TRP1 gene, and profile by gas chromatography of the total sterols that they produce. ;For this, the 41 colonies were transferred to a minimal medium containing respectively 10, 20 or 50 ug/ml nystatin; about ten colonies are able to grow on the medium containing a dose of 50 ug/ml nystatin. These resistant colonies are selected from verifying their gene structure and also their sterol compositions. ;The insertion of the TRP1 gene into the ERG6 geneX is verified by PCR using a pair of oligonucleotides covering the junction between the functional TRP1 gene and the broken ERG6. This pair of oligonucleotides is as follows: OERG6trp3: AGGGCCGAACAAAGCCCCGATCTTC ;(SEQ ID No. 3) and OERG6trp4: GGCAAACCGAGGAACTCTTGG (SEQ ID No. 4). ;Some strains exhibit the expected PCR profile, that is, a fragment of 800 base pairs corresponding to the size expected for a ZKPi insertion into ERG6. With the intention of verifying that ERG6-geneX is actually inactivated in these strains, an analysis of the sterol compositions of these strains using gas chromatography and high-pressure liquid chromatography was carried out (Duport et al., 2003; Szczebara et al., 2003). ;These analyzes confirm the absence of ergosterol synthesis and the accumulation of abnormal sterols compatible with the expected disruptions of the biosynthetic pathway in the disrupted strain. ;One strain was more definitely chosen, and is called WGIFOl. ;Example 2: Construction of CA10, CA14 and CA23 strains. ;CAlO strain (with genotype): MATa, rho<+>, GAL2, ura3- 52, trpl- A63, his3- A200, ergSr. HYGRCP, ade2::GAL10/ CYCI::A7Reductase::PGKl, LEU2::GAL10/ CYCI::matADR::PGKl), CA14 strain (of genotype: MATa, rho<+>, GAL2, ura3- 52, trpl - A63, his3- A200, ergSr.HYGRO* atf2::G418<R>, ade2::GAL10/ CYCl::A7Reductase::PGKl, LEU2::GAL10/ CYCl::matADR::PGKl), and the CA23 strain (by genotype: MATa, rho<+>, GAL2, ura3- 52, trpl- A63, his3- A200, erg5::HYGRO^, arel::G418<R>, are2::HIS3, ade2::GAL10/ CYCl ::A7Reductase::PGKl, LEU2::GAL10/ CYCl::matADR::PGKl), and also the constructions thereof, are described in the reference Duport et al., the technical content of which, with respect to the construction of these strains is incorporated into the present application by reference.

These strains produce and contain, in their membranes, unnatural sterols (as described in European patent application EP0727 489) and especially ergosta-5-enol (campesterol).

These three strains do not express the product of the £i?G5 gene, which is non-functional due to the insertion into its coding sequence of the hygromycin resistance gene. In addition, these strains express cDNA encoding plant 87-reductase (European patent application EP 0727 489 specifically describes the cloning of the 87-reductase from the plant Arabidopsis thaliana, which is incorporated into the present application by reference, the Genbank accession number of this sequence is ATU49398).

The CA14 strain is derived from the CA10 strain by disruption of the ^4ZF2 gene. The product of this gene results in the acetylation of pregnenolone at position 3 (as described in patent application WO99/40203).

The CA23 strain is a strain derived from the CAlO strain by deletion of ARE 1 and ARE2 genes, the two proteins Are lp and Are2p are responsible for the esterification of ergosterol (Sturley, 2000) and possibly of cholesterol because they are homologous to the enzyme that is responsible for the esterification of cholesterol in mammals

(AC AT).

Example 3: Construction of the plasmid to express the 524-25 reductase of human origin

(plasmid pYES_Delta24)

The construction of this plasmid was described by Waterham et al., 2001. The construction consisted of placing the cDNA encoding delta 24 sterol reductase under the control of the pGALl promoter and of the tCYCl terminator in the vector pYES2 (Invitrogen SARL, Cergy Pontoise, France ). This plasmid is an ice cream. coli/ S cerevisiae exchange plasmid and contains a 2 micron origin of replication and a URA3 g& a., which allows it to replicate in yeast and makes it easy to select the yeast transformed with this plasmid.

In addition, the GALI promoter is galactose inducible.

Example 4: Construction of the plasmid pAG1 expressing A. thaliana expressing 87-reductase

A plasmid was constructed specifically for the expression of A. thaliana delta 7 reductase, on a single copy vector.

The plasmid pAM1 was used for this construction. This plasmid, the construction of which is described in PCT application WO 02/061109 (cf. example 9.b in said application, which is incorporated in the present application by reference), is an ice. coli/ S. cerevisia exchange plasmid based on an autonomous replication sequence and a centromere (ARS CEN). The selection marker is the ADE2 gene. This plasmid is compatible, and therefore can replicate at the same time as a plasmid based on a 2 micron origin of replication. In particular, this plasmid has a unique NotI site for cloning expression facets, as described in the PCT application above.

This site was used to clone an A. thaliana Delta7 reductase expression cassette derived from the CALO strain. Indeed, this expression cassette is very efficient and enables the CAlO strain, which is also disrupted in ERG5, to produce campesterol (ergosta-5-enol) as the major sterol (Duport et al., 1998). The genomic DNA fragment of the CAlO strain containing the delta7 reductase gene is transformed using the following primers:

The transformation is performed on the genomic DNA of the CAlO strain made by the rapid phenol/chloroform extraction technique as described by Adams et al. (Adams and Holm, 1996).

Fifty nanograms of CA10 genomic DNA was used as matrix for transformation using primers OSA72 and OSA77. The Taq DNA polymerase and the enzymatic conditions were from the company Stratagene. The denaturation conditions were as follows: initial denaturation for 5 minutes at 95°C, then 30 cycles were performed consisting of a denaturation for 30 seconds at 95°C, an annealing for 30 seconds at 50°C and then an elongation for 1 minute at 72°C. The reaction is terminated by a final extension of 10 minutes at 72°C.

The PCR fragment was then cut with the NotI enzyme and purified on agarose gel and then conventionally cloned into the unique NotI site of the plasmid pAM1. The plasmid thus obtained was named pAG1.

It is a single copy vector for the expression, thaliana delta7 reductase in yeast, the delta7 reductase gene is placed under the control of the GAL 10/CYCl promoter (Lecain et al., 1996).

Example 5: Extraction of the free and esterified sterols in iron, for the analyses:

1) Conditions for extracting the free and esterified sterols in yeast (procedure 1).

a) Conditions for extracting the free sterols:

The cell pellets are washed twice with 500 µl of deionized and filtered water, in a glass tube.

The cells are then resuspended in 500 µl of water containing 0.5 mm glass beads, which corresponds to 150 µl of liquid in the tube.

An extraction is carried out twice with 2 ml of 1,2-dichloroethane with vigorous agitation on a vortex for 10 minutes. After the first extraction, the mixture of cells, glass beads and solvent was centrifuged for 5 minutes at 1500 g in order to separate the two phases.

The two organic fractions derived from the two subsequent extractions are combined and dried under a stream of nitrogen for a few hours.

The sterol extract is suspended in 100 µl acetonitrile for high performance liquid chromatography (HPLC) analysis (Szczebara et al., 2003) or in 100 µl hexane for gas chromatography (GC) analyzes (Duport et al., 2003). b) Conditions for extracting the total sterols: saponification and extraction of the esterified sterols, qualitative analysis procedure 1: The cell pellet is resuspended in 500 ul of purified water. 2 ml of potassium hydroxide KOH at 10% in methanol is added to this suspension. The mixture is heated for 1 hour at 60°C in closed tubes. After the incubation, and once the tubes have returned to ambient temperature, the mixture is extracted three times with 2 ml of hexane. Between each extraction, the two phases are separated by centrifugation for 5 minutes at 1,500 g. After each extraction, the organic phase is transferred to a new tube, and the three organic phases are then combined and then dried under a stream of nitrogen.

Sterol DNN is resuspended in 100 µl of 100% acetonitrile for analysis by high-performance liquid chromatography (HPLC) (Szczebara et al., 2003) or in 100 µl of hexane for the gas chromatography (GC) analyzes (Duport et al., 2003 ). 2) Conditions for extracting the free and esterified sterols in the yeast, for qualitative analysis (procedure 2).

The strains are grown in rich medium (10 g bactopeptone per liter and 10 g yeast extracts per liter) with 2% glucose as carbon source to provide 500 mg of lyophilized cells. These dried cells are taken up in 3 ml of methanol (100%) containing 1 g of KOH and a table of pyrogallol, and the mixture is then incubated for 45 minutes at 90°C. After returning to ambient temperature, the sterols are extracted with 5 ml of hexane. The organic phase was separated into three samples having the same volume, and is dried under a stream of air. Two of the samples of the extracted sterols were taken up in 100 µl of hexane for the analyzes by gas chromatography (GC) and gas chromatography/mass spectrometry GC/MS, and the third sample was taken up in

150 µl methanol for the high performance liquid chromatography (HPLC) studies.

Example 6: Analysis of the free and esterified sterols in the yeast using gas chromatography

(GC).

1) Gas chromatography (GC) with FID (flame ionization detection).

The sterol extract (free or total) suspended in hexane is prepared according to procedure 1 (cf. example 51) a) and b)). An injection control is added to the sterol mixture, usually cholesterol at a concentration of 10 to 50 ng/µl.

From 1 to 3 µl of samples are then injected onto a gas chromatography device under the following conditions. 1 to 3 µl was injected onto an Alltech SE30-type column (column reference: 20 m x 0.32 mm IDX 0.25 µm). The gas vector is helium. The split ratio is between 50 and 80. The column head pressure is 30 psi. The injector is set at 280°C. The initial temperature of the column is 130°C for half a minute. It increases to 230°C at a rate of 40°C per minute, and then from 230°C to 280°C at a rate of 3°C per minute. The column was maintained at 290°C. The temperature of the detector is 310°C. 2) Gas chromatography (GC) with FID (flame ionization detection) coupled with mass spectrometry

(GC/MS).

The total sterol extract suspended in hexane is prepared according to procedure 2. The GS used is equipped with a conventional "split/splitless" injector with a conventional DB5 column 30 meters in length and 0.25 mm in diameter.

The injection is carried out at 230°C with helium as gas vector, at a flow rate of 2 ml/min. The column runs from 130-290°C in 4 steps. The column is maintained at 130°C before injection and then increases to 230°C with a ramp of 40°C per minute, then from 230°C to 280°C with a ramp of 3°C per minute and then from 280°C to 290°C with a ramp-up of 30°C per minute. The column remains at 290°C for 5 minutes.

At the gas chromatography column outlet, the molecules are then analyzed by spray mass spectrometry in an ionization chamber such as that of a Perkin Eimer turbo mass type device. The molecules were fragmented with a high energy electron beam. The different fragments were then separated on a quadrupole filter and then detected on an ion detector. A mass spectrum that includes the masses of all products from fragmentation of the M+ ion corresponds to each mass located on the ion current graph. This mass spectrum obtained at a given retention time on the column is compared with the libraries for fragmented products and also with those described for sterols by Quail and Kelly. (Methods in Molecular Biology Vol. 53 Yeast Protocols Edited by Evans; M. Quail and S. Kelly "The Extraction and Analysis of Sterols from Yeast" pp. 123-131 (1996)).

In this way it was possible to demonstrate the effect of the deletion of the ERG6 gene in the WGIFOl strain, and in particular the absence of ergosta-8, 24 (28)-dienol and the presence of cholesta-type sterol which has the double bond in 24 (25)- the position.

Example 7: Analysis of the free and esterified sterols in the yeast using high performance liquid chromatography (HPLC) with UV detection or detection using mass spectrometry.

1) Analysis by UV detection HPLC:

10 to 30 µl of the sterol extract (suspended in acetonitrile or methanol and prepared according to procedure 1 or 2 (cf. Example 5) is injected onto em X terra RP18 type 4.6 x 100 mm column (Waters, Milford, MA01757 USA).

The separation is carried out on a gradient composed of water containing 0.02% TFA (trifluoroacetic acid) (buffer A) and of pure acetonitrile (buffer B). The column is maintained at 60°C during the analysis.

The HPLC device used is of the "Waters 600E System Controller" type (Waters, Milford, MA01757 USA). The UV detection was performed on a diode array detector covering the wavelengths from 206 to 350 nm. The column was equilibrated with a buffer containing 20% (v/v) of buffer A (acetonitrile) and 80% of buffer B (water containing 0.02% TFA (trifluoroacetic acid)). A linear gradient is formed from a solution containing 50% buffer A and 50% buffer B. After 10 minutes, the composition of the elution buffer becomes 25% of buffer A for 75% buffer B. A new linear gradient is then used in such a way that at 30 minutes the gradient reaches the value of 100% of buffer B. This value is maintained for 5 minutes to clean the column.

2) Analysis by HPLC, detection by mass spectrometry (HPLC/MS):

In the case of an analysis by mass spectrometry, the sample is maintained at 30°C and the column is maintained at 60°C during the analysis. The HPLC device used is of the "Allicante HT Waters 2790" type, connected to a "Waters MicroMass ZQ" mass detector. Unlike the previous detection method, elution buffer A does not contain TFA, but the two buffers A and B contain 0.01% (v/v) formic acid.

The column was equilibrated with a buffer containing 80% of buffer A' (water containing 0.01%

(v/v) of formic acid) and 20% buffer B' (acetonitrile containing 0.01% (v/v) formic acid).

The injection begins with a buffer containing 50% of these two buffers. A linear gradient with two runs is formed from a solution containing 50% buffer A' and 50% buffer B'.

After 10 minutes, the composition of the elution buffer is 25% of buffer A' and 75% of buffer B'. The run-up of the gradient was then modified so that it reaches 12.5% of buffer A' and 87.5% of buffer B' after analysis for 25 minutes, and then 100% of buffer B at 30 minutes. This value is maintained in 5

minutes to regenerate the column.

The "Waters MicroMass ZQ" mass detector is set for positive electrospray in ionization scanning. The value of m/z is between 295 and 450. A "continuum" approach is selected for the scan. Furthermore, a signal extraction in the "SIR" mode is performed in parallel at all the expected masses in natural isotope presence for the sterols to be analyzed. The detector is set so that it is capable of completely resolving without interference between molecules that differ by one unit of m/z. All the accesses are parameterized in such a way that the total access time corresponds to the scan, and the total time for access of all the SIRs remains less than 2 seconds.

Example 8: Cultivating in the yeast strains to analyze their sterol content with or without<13>C labeling: The strains to be analyzed were grown in a volume of 50 ml Kappeli medium (Kappeli et al., 1985) containing 2% normal D- glucose or D-glucose-U-<13>C6 (for the labeling experiments, cf. Figure 10).

The optical density of the starter culture is 0.1 at 600 nm. This culture is incubated for 72 hours at a temperature of 30°C with shaking at 200 rpm.

The cells were then recovered by centrifugation of the medium at 600 g for 10 minutes. The cell pellet was then analyzed directly by the analytical techniques presented in Example 5 (preliminary study and which does not require induction with galactose).

However, for the studies regarding the kinetics of induction of the expression of delta 7-reductase and of delta 24-reductase (strains transformed with the plasmid pYES_Delta24 and/or pAGl), the pellet is resuspended in 50 ml of fresh Kappel medium containing 2% galactose (unlabeled with<13>C carbon).

This culture is incubated at a temperature of 30°C with shaking at 200 rpm.

Ten ml of culture is recovered after 0 hours, 2 hours, 4 hours, 8 hours and 24 hours of cultivation.

These culture samples are centrifuged at 800 g for 10 minutes, and the cell pellet is frozen and stored at -20°C before sterol injection using the methods described in example 5.

Example 9: Identification of the sterols present in the strains analyzed: The identification of the sterols is based on the combination of the following principles: Comparison of the behavior by GC, HPLC, GC/MS and HPLC/MS with authentic standards in the case of campesterol (ergosta-5-enol), of ergosterol (ergosta-5, 7, 22-trienol) of cholesterol (kolesta-5-enol), of desmosterol (kolesta-5,24-dienol) of cholesta- 5, 22-dienol and of zymosterol (cholesta-8, 24-dienol).

Analysis of absorption spectra by means of HPLC and diode-matised UV detection (cf. example 7-1)): This procedure makes it possible to identify unequivocally, on a spectral basis, five classes of sterols: 1) class SA1: non-conjugated diene system, 2) class SA2: presence of a 5,7-diene system; 3) class SA3: presence of a 22, 24 (25)-diene system; 4) class SA4: presence of an 8,14-diene system; 5) class SA5: presence of a 22, 24 (28)-diene system. Classes SA3 and SA5 cannot co-exist due to structural reasons. Classes SA2 and SA4 cannot coexist due to biosynthetic reasons, class SA2 can be combined with structural units of classes SA1, SA3, SA5 to form

additional composition spectra.

Analysis of the retention times in GC and in HPLC on the basis of an approximate additivity of the retention time changes associated with each type of unsaturation and with the presence of an ergosta- or cholesta-type backbone. Because this criterion is not absolute, it is used as an aid to identify and to remove ambiguities, but presents a risk of error if used alone. It is therefore only used in combination with the other criteria.

GC/MS analyzes (cf. Example 6-2)), which provide the molecular mass and a

fragmentation profile comparable to spectral libraries.

HPLC/electrospray MS analysis (cf. Example 7-2)), which provides, in the case of 3-hydroxysterols, a major signal at the molecular mass -17 (protonation (+1) and loss of a water molecule (-18)) .

Analyzes with all the above-mentioned systems of the sterol compositions with different

reference yeast strains that are disrupted at different points in biosynthesis.

- Analyzes of variations in the sterol composition during complementation with different biosynthetic enzymes, and of the kinetics of this complementation during the induction of this complementation.

Analyzes of the labeling profile of the different sterols with the carbon-13 isotype.

- Analyzes of the UV spectrum of the sterols separated by HPLC at a given retention time. The two 5,7-conjugated double bonds exhibit a typical spectrum with two absorption peaks between 265 and 280 nm, while the two 22,24-conjugated double bonds exhibit an absorption peak at 235 nm. The final, 8,14-conjugated double bonds can be identified by an absorption peak at 245 nm.

Example 10: Identification of the sterols present in the BMA64 strain:

The BMA64 strain is grown in a volume of 50 ml Kappeli medium containing 2% D-glucose for quantitative and comparative analysis of the sterols.

The optical density of the starter culture is 0.1 at 600 nm. This culture is incubated for 72 hours at a temperature of 30°C with shaking at 200 rpm.

The cells are then recovered by centrifugation of the medium at 600 g for 10 minutes, and the cell pellet is analyzed by the techniques presented in Example 5. The various analyzes described make it possible to identify the sterols produced by this strain.

It was thus determined that this strain accumulates more than 80% of its free sterols in the form of ergosterol (ergosta-5, 7, 22-trienol) (cf. Figure 8). Two other less detectable sterols were produced by this strain; they are ergosta-5,7-dienol (substrate for the product of the ERG5 gene) (12%) and zymosterol (ergosta-8, 24-dienol) (5%). No trace of cholesterol is detectable (the limit of detection in the method is approx. 0.5% of observable sterols). Small amounts of lanosterol are also detectable (only in the analyzes of total sterols).

Example 11: Identification of the sterols present in the WGIFOl strain:

The WGIFOl strain (cf. Example 1) was grown in a volume of 50 ml of Kappeli medium (Kappeli et al., 1985) containing 2% D-glucose for the quantitative and comparative analysis of the sterols.

The optical density of the starter culture is 0.1 at 600 nm. This culture is incubated for 72 hours at a temperature of 30°C with shaking at 200 rpm.

The cells are then recovered by centrifugation of the medium at 600 g for 10 minutes, and the cell pellet is analyzed by the techniques presented in Example 5. The various analyzes described make it possible to identify the sterols produced by this strain.

The search, in the chromatogram, for ergosta-5, 7, 22-trienol (ergosterol) or for ergosta-5, 7-dienol is negative (less than 0.5% of the value obtained in BMA64) in HPLC coupled to mass spectrometer. Regarding the free sterols, the strain accumulates 50% of the total zymosterol (cholesta-8, 24-dienol), substrate for the product of the ERG6 gene and 30 and 20% respectively of cholesta-5, 7, 24-trienol and of cholesta -5, 7, 22, 24-tetraenol possibly resulting from a mechanism of synthesis identical to that resulting in ergosta-5, 7-trienol in ergosta-5, 7, 22-trienol in the parent strain (compare Figs 3 and 8). This clearly shows that the biosynthetic reaction pathway is blocked at the level of erg6 because the is7?G<5p enzyme (S-adenosylmethionine delta24sterol C-methyl-transferase) converts cholesta-8, 24 (25)-dienol to ergosta-8, 24 (28 )-dienol (cf. Figure 2). This accumulation clearly indicates that the WGIFOl strain does not possess a functional copy of the erg6 gene. The results also indicate that the normal ergosterol biosynthetic pathway in yeast, and in particular sterol 8,7-isomerase, sterol 5-desaturase and sterol delta 22-desaturase, is able to convert the cholesta-type substrates with an activity that remains significant.

Example 12: Contraction of the WGIF02 strain and identification of the sterols present in this strain: The WGIF02 strain was obtained by transforming the WGIFO1 strain with the plasmid pYES2 carrying the 824 reductase expression cassette (pYES_Delta24, cf. Example 3). The clones were selected on a medium lacking uracil and the presence and expression of 524-reductase cDNA verified by analyzing the sterols of these transformants using procedure 1 (cf. Example 5-1)).

A clone named WGIF02 was selected because it had a sterol profile different from the WGIFO1 strain; furthermore, the additional sterol had a retention time similar to that of cholesterol (cf. Figure 7). The WGIF02 strain is grown in a volume of 50 ml Kappeli medium (Kappeli et al., 1985) containing 2% D-glucose, for the quantitative and comparative analysis of the sterols.

The optical density of the starter culture is 0.1 at 600 nm. This culture is incubated for 72 hours at a temperature of 30°C with shaking at 200 rpm.

The cells were then recovered by centrifugation at 600 g for 10 minutes, and the cell pellet analyzed by the techniques presented in Example 5. The various analyzes described make it possible to identify the sterols produced by this strain.

The two sterol extraction profiles for the WGIFOl strain and for the WGIF02 strain are similar except for the presence of a new peak that was identified, by its mass, its retention time and its conjugated double bonds, as cholesta 5,7,22-trienol (Figures 2, 3 and 7). The presence of this compound indicates the expected presence of a 24, 25-sterol reductase activity on the double bond in the 24(25) chain of cholesta-5, 7, 22, 24(25)-tetraenol. In addition, the amount of cholesta-5,7,24-trienol is reduced by the presence of cholesta-5,7,22-trienol in the WGIF02 strain (Figures 7 and 8). The activity of the enzyme is shown by the conversion of cholesta-5, 7, 24-trienol which represents 30% in the WGIFOl strain and around 12% in WGIF02, the difference, that is 18%, is completely in the form of cholesta-5, 7, 22-trienol in the WGIF02 strain. This is an unexpected result since the product of conversion of cholesta 5,7,24 by the delta 24 reductase is cholesta-5,7, which is absent in WGIF02, and therefore quantitatively converted to cholesta-5,7,22. This shows another unexpected result, that is, cholesta 5,7 is a substrate for strol 22-desaturase, while cholesta 5,7,24 is, according to the sterol profile of the WGIFOl strain, a poor substrate.

Example 13: Construction of the WGIF03 strain and identification of the sterols present in this strain: The WGIF03 strain was obtained by transforming the WGIFO1 strain with the plasmid pAG1. This exchange plasmid between E. coli and S. cerevisiae carries an expression cassette for 87-reductase, the corresponding cDNA of which is under the control of the GAL 10/CYCl promoter. The WGIFO1 strain was transformed by the lithium chloride technique, and the transformants were selected on medium containing adenine. The expression of delta7 reductase was verified by the presence, in the sterol profile of the correct clones, of cholesta-5, 24 (25)-dienol. A clone satisfying these criteria was more specifically selected and named WGIF03.

The WGIF03 strain was grown in a volume of 50 ml Kappeli medium (Kappeli et al, 1985) containing 2% D-glucose, for the quantitative and comparative analysis of the sterols.

The optical density of the starter culture is 0.1 at 600 nm. This culture is incubated for 72 hours at a temperature of 30°C with shaking at 200 rpm.

The cells are then recovered by centrifugation of the medium at 600 g for 10 minutes, and the cell pellet is analyzed by the techniques presented in Example 5. The various analyzes described make it possible to identify the sterols produced by this strain.

The expression of the delta7 sterol reductase in the WGIFOl strain (to give the WGIF03 strain), unlike the expression of the delta24 sterol reductase, results in a pronounced change in the sterol profile of the strain, with an almost complete absence of cholesta-5, 7, 22 , 24-tetraenol, cholesta-7, 24-dienol and cholesta-8, 24-dienol.

This activity is also marked by the presence of a major peak identified as previously described as cholesta-5, 24-dienol or desmosterol. The amounts of cholesta-8, 24-dienol change from 12 to 48%. The detected cholesta-5,7,24-trienol shifts from 30 to 3% and the detected cholesta-5,7,22,24-tetraenol shifts from 23 to 4%, respectively, for the WGIFO1 and WGIF03 strains. These observations indicate, unexpectedly, that the sterol delta7 reductase reduces cholesta-5,7-dienol in a manner that is, so to speak, dependent on the nature of the unsaturations carried by the side chain of the sterols. This result is opposite to that observed with the sterol delta24 reductase. Unexpectedly, the expression of the sterol delta7-reductase also results in the accumulation (12%) of a molecule that co-migrates with the cholesta-5, 7-dienol. However, it is found to be relatively unlikely, although not impossible, that this molecule is cholesta-5,7-dienol, the theoretical level of which should decrease and not increase under these conditions. The presence of a small amount of cholesta-5, 7, 24-trienol (8%) is also of interest. The latter sterol is the unexpected product of the action of sterol 22-desaturase on cholesta-5, 24-dienol, the major sterol (60%) in the WGIF03 strain (which results from the reduction of cholesta-5, 22, 24-trienol by sterol delta7 reductase). The small accumulation of cholesta-5,22,24-trienol indicates, unexpectedly, that cholesta-5,24-dienol is not a good substrate for sterol 22-desaturase. In light of the results obtained in the WGIF02 strain (cf. Example 12), it can be deduced that the presence of an unsaturation in the 24-position (cholesta-5, 24-dioenol or cholesta-5, 7, 24-trienol) makes the difficulty for the sterol 22-desaturase to metabolize the sterols. The action of the ERG6p enzyme in converting cholestas' unsaturation at the 24-position to ergosta therefore results in the conversion of poor substrates for the ERG5 gene (22-desaturase) into good substrates.

Example 14: Construction of the WGIF04 strain and the identification of the sterols present in this strain: The WGIF04 strain was obtained by transforming the WGIF02 strain with the plasmid pAG1 by the lithium chloride technique, and the transformants were selected on medium containing neither adenine nor uracil. The correct transformants were then confirmed on the basis of the detection of cholesterol accumulation. A clone that satisfied these criteria was more specifically selected and named WGIF04. A sample of the WGIF04 strain was deposited in the Collection Nationale de Cultures de Microorganismes (CNCM) [National Collection of Microorganism Cultures], Institut Pasteur, 25, rue du Docteur Roux, 75724 Paris Cedex 15, France, on April 22, 2004 under the registration number 1-3203.

Strains indistinguishable from WGIF04 can also be obtained by transforming the WGIF3 strain with pYES_Delta24 and using the same selection.

The WGIF04 strain was grown in a volume of 50 ml in Kappeli medium containing 2% D-glucose, for the quantitative and comparative analysis of the sterols.

The optical density of the starter culture is 0.1 at 600 nm. This culture is incubated for 72 hours at a temperature of 30°C with shaking at 200 rpm.

The cells are then recovered by centrifugation of the medium at 600 g for 10 minutes, and the cell pellet is analyzed by the techniques presented in Example 5. The various analyzes described make it possible to identify the sterols produced by this strain.

Cholesterol represents 25% of the free sterols in the WGIF04 strain (cf. Figure 9). The formation of cholesterol in this strain has been independently shown in GC and in HPLC, by simultaneous migration with an authentic standard, and by confirmation both in GC/MS and in HPLC/MS. Cholesterol is not detectable (< 0.5% of total sterols) in all strains that do not simultaneously express delta7-reductase and delta24-reductase.

The strains that do not have a deletion in the ergtf gene can produce cholesterol; however, this represents less than 5% of the total free sterols. It was thus possible to construct the BMA64-pYES_Delta24-pAG1 strain obtained from BMA64 simultaneously transformed with pYES_Delta24 and pAG1. This strain produces cholesterol, the latter representing a few % of the total sterols.

The CAlO strain was transformed with the plasmid pYES_Delta24. This strain also produces cholesterol, the latter representing a few % of the total sterols.

It was further possible to demonstrate that the formation of the cholesterol clearly requires the induction of the delta7-reductase and delta24-reductase promoters (compare Figures 4 and 5). The strains containing these genes do not produce cholesterol in the absence of induction; figure 5 indicates that the maximum level for cholesterol is reached after approximately 24 hours of induction. In parallel with the formation of cholesterol (cholesta-5-enol), the formation of cholesta-5, 22-dienol is also observed. Analysis of Figure 4 and Figure 5 indicates that the formation of the latter compound takes place faster after an induction than the formation of cholesterol, and even begins before induction (cf. Figure 5A: m/z = 367). However, this connection is completely absent if the strain does not carry the two plasmids pAG1 and pYES_Delta24. The formation of 22-dehydrocholesterol is therefore a much faster process than that of cholesterol formation, but this process involves a precursor which disappears quickly after induction, leaving room for the formation of cholesterol. The cholesterol can be formed from cholesta-5, 24 via the 824-reductase or from choleta-5, 7 via the 87-reductase. Now it has been shown that cholesta-5, 7-dienol cannot accumulate due to the fact that it is instantly converted into cholesta-5, 7, 22-trienol. The source of cholesterol is therefore cholesta-5, 24-dienol, which is absent at the time of induction and accumulates at around 4-8 hours of induction, before decreasing at around 24 hours (Figure 5). This explains the late appearance of cholesterol because synthesis of cholesta-5, 24-dienol is required beforehand. Conversely, two possible precursors of cholesta are 5,22-dienol, cholesta-5,7,22-trienol and cholesta-5,22,24-trienol. The latter is absent at the beginning of induction (Figure 4) while the former is present and then rapidly decreases in parallel with the standardization of the formation of cholesta-5, 22-dienol. It can be concluded for this that the source of cholesterol is the reduction of 5, 24-dienol by the 824-reductase, while the formation of cholesta-5, 22-dienol results from the reduction of 5, 7, 22-trienol by the 87-reductase. The formation of cholesta-5, 22-dienol by the action of 822-desaturase for cholesterol is not completely ruled out, but appears to be a minor process on the basis of the preferential accumulation of cholesterol compared to cholesta-5, 22-dienol at the long-term point (24 hours) in the kinetics (Figures 4 and 5).

Example 15: Optimization of the cholesterol biosynthetic pathway, the role of 822-desaturase: For induction times up to 24 h for the WGIF04 strain, the accumulation of cholesta-5, 22-dienol represents about 50% of that of cholesterol with respect to free cholesterols ( figure 4). Disruptions of the 822 desaturase gene are an option to optimize cholesterol production. The construction of a strain doubly disrupted in the 822-desaturase (erg5 gene) and in the ergtf gene and expressing 87-reductase and 824-reductase is entirely conceivable. A strain carrying the subset: increase of 822-desaturase, the expression of 87-reductase and the expression of 824-reductase was produced.

This strain was obtained by transforming the CAlO strain with the plasmid pYES_Delta24 using the lithium chloride technique and selecting for prototrophy with respect to uracil. The strain obtained was named CalO/824.

The CAlO strain expressing 824-sterol reductase produces a relatively small amount of cholesterol (compare Figures 6 and 7) and accumulates mainly ergosta-5-enol and an intermediate amount of ergosta-5,7-dienol. The accumulation of cholesta-5,7-dienol is very low in such a strain, indicating that disruptions of the erg6 gene are essential for substantial accumulation of their derivatives of the cholesta series. The activity of the 824 reductase therefore competes, surprisingly enough, relatively little with that of the product of the ergtf gene. It can therefore be concluded from these results that the simultaneous destruction of the erg5 and ergtf genes is important for the purpose of optimizing the production of cholesterol.

The destruction of the erg5 gene in the WGIF04 strain would enable those skilled in the art to significantly increase the production of cholesterol. The possibility of such a strain is established by the construction of the WGIF04 and CA10/524 strains and the fact that this strain is only a natural genetic combination that can be produced from the two previous strains. The data derived from WGIF04 indicate that cholesta-5, 24 disappears rapidly when 524 reductase is expressed (by comparing the results obtained with WGIF03 and those obtained with WGIF04). This makes it possible to predict that cholesterol will be synthesized very efficiently in a strain that is simultaneously destroyed in the erg5 and ergtf genes, and at the same time expresses 87-reductase and 824-reductase, then cholesterol becomes the only final sterol.

In conclusion, the minimum requirement for the production of cholesterol is a threshold above or equal to 20% of the total sterols, destruction of the ergtf gene, and expression of the 87-reductase and of the 824-reductase. Complementarity destructions of 822-desaturase would make it possible to improve cholesterol productivity and to eliminate the parasitic formation of cholesta-5, 22-dienol as the final sterol.

Example 16: Isotope labeling of cholesterol and definition of isotope signatures:

The principle for the production of a labeled cholesterol is described in Figure 10. This manipulation consists first of all in growing the yeast on glucose completely labeled with<13>C, for 72 hours with a cultivation at 30°C.

The cells are then recovered by centrifugation of the medium at 600 g for 10 minutes. The cell pellet is then suspended in 50 ml of fresh Kapelli medium containing 2% galactose which is not labeled with <13>C-carbon. The cultures are stopped 2 hours, 4 hours, 8 hours or 24 hours after the shift to galactose, and the sterols are extracted and then analyzed (cf. Example 7). The shift from glucose to galactose results in the induction of GAL 10/CYCl promoters that control both the 87 reductase gene and the 824 reductase gene. At the same time, there is a change in the isotopic labeling of the carbon source. There follows the de novo synthesis of metabolic intermediates and then of sterol, including cholesterol, which includes a gradual change in labeling. This change in labeling can be characterized by the mass profile of each intermediate sterol. In fact, the incorporation of each<13>C atom induces a change in mass of one atomic mass unit (AMU). The cholesterol, for example, thus appears to have a molar mass in the range from 386 to 413 daltons, depending on the degree of labelling. In analysis by HPLC mass, using positive electrospray ionization, this corresponds to m/z (mass/charge) values in the range from 369 to 396 (ion M+ H+ - H2O, i.e. M+l-18 = 385-17 = 369). Sterol retention time in HPLC is not dependent in a detectable way on the degree of labelling. The mass spectrum of an HPLC peak corresponding to a single sterol "X" corresponds to a mass distribution which is therefore the superposition (sum) of the mass distributions of sterol "X" synthesized at the various times after labeling. It is therefore a profile that is complex (Figure 11), but can be fully experimentally determined and that represents a unique isotopic signature that depends simultaneously on: 1) the labeling protocol, and especially on the culture times, and conditions with<12>C-glucose and with<13>C-galactose,

2) the exact genetic makeup of the strain used;

3) the exact time at which the cultures are stopped.

This isotope profile has several unique properties:

1) it can be modulated as desired by adjusting cultivation conditions, the strain used and the sterol chosen. A unique marking register can therefore be produced; 2) it is capable of combining, that is, multiple isotopic signatures corresponding to multiple unique sterols labeled with isotopic profiles that are themselves modifiable can be combined to form a "molecular" alphabet; 3) it is reproducible, and it is easy to determine experimentally (cf. Figures 11 and 12) the double or triple pots indicating the reproducibility of the profiles; 4) it corresponds to a molecular tracer mixture that is easy to isolate, stable, colorless and odorless, non-volatile and non-toxic, and it can be incorporated into food, a medicinal product, additives or other products that can be assimilated by humans; 5) it cannot be faked without having the specific recombinant strains and the very precise labelling, cultivation and extraction conditions. In addition, knowledge of the isotopic signature does not make it possible to trace back to the parameters that made it possible to produce it.

In summary, a general-purpose "isotope alphabet" that cannot be falsified and that can be incorporated into products of any type, including food, can be readily provided by the

the present invention. There are actually virtually an unlimited number of "isotopes" that can be constituted from such an alphabet by making use of both the labeling profiles and the different types of sterols. The incorporation of such signatures into the most varied products therefore constitutes a unique method of marking which cannot be forged, unlike for example DNA signatures which can be reproduced as soon as they are known. The signature can also be read non-destructively, for example by laser ionisation followed by mass spectrometry analysis (MALDI-TOF or similar).

Example 17: Production of non-animal cholesterol strongly labeled with <13>C:

The use of<13>C-galactose or of<13>C-ethanol and glucose instead of the unlabeled carbon sources to grow WGIF04 strains under the conditions described above for the sterol assays makes it possible to synthesize very strongly labeled sterols and especially cholesterol ( which comprises at least 95% of<13>C-carbon). The preparation of<14>C-radioactive sterols and cholesterol is also possible using the same approach. The method can also be incorporated into yeast strains that produce steroids, and especially hydrocortisone (cf. patent application WO 02/061109) to produce<13>C-labeled or<14>C-labeled steroids, for example for RIA assays.

Example 18: Contraction of a strain that produces mainly cholesterol:

The CDR07 Mata strain is described in the patent application published under the number WO 02/061109 and deposited with the Collection Nationale de Cultures de Microorganismes (CNCM) [National Collection of Microorganism Cultures], rue du Docteur Roux, 75724 Paris Cedex 15, France , according to the provisions of the Treaty of Budapest at the CNCM, 24 January 2001, under registration number 1-2616.

The CDR07 MATa strain (relevant markers: ade2::GAL10/ CYClp::A7; erg5::PGKlp::hygro<R>; ERG6) was crossed with the WGIFOl strain described in Example 1 (relative markers: ERG5; erg6 ::TRPl).

After sporulation of the diploid, the sterol composition of the spores was determined to find spores that produce desmosterol, which is the precursor of cholesterol, as described in Example 6. A spore that produces desmosterol, which has a functional TRP1-gea. and carrying the Arabidopsis thaliana 87-reductase cDNA, as indicated by the PCR analysis, was thus identified. In addition, this strain, called YIM59, is sensitive to hygromycin, indicating that the ii7?G5 gene is functional.

A sterol preparation, made as described in Examples 6 and 9, showed that this YIM59 strain produces sterols having the same retention time as zymosterol and desmosterol. The YIM59 strain also showed auxotrophies for adenine, leucine, uracil and histidine. The presence of desmosterol shows that this strain expresses the sterol 87 reductase and that it carries a non-functional erg6 allele.

In order to improve the level of expression of human DHCR24, the nucleotide sequence of DHCR24 cDNA was modified; to enhance the translation of the initiating ATG, the promoter controlling the expression of DHCR24 was also modified. The new promoter chosen is the CTCi promoter of cytochrome c1, as a replacement for the G4L/-inducible promoter of the plasmid pYES Delta 24 (cf. Example 3).

The sequence corresponding to the terminal NH2 of the DHCR24 reductase as a fusion with the CYC2 promoter was modified as follows: tagcgtggatggccaggcaactttagtgctgacacatacaggcatatatatatgtgtgcgacgacacatgatcatatggcatgcatgtgctctgtatgtatø aactcttgttttctttttctctaaatattc gtacaaaaaagcaggctaaaaaATGGAACCTGCCGTGTCGCTGGCCGTGTGCG (SEQ ID NO. 7).

The lowercase letters represent the partial nucleotide sequence of the CYC1 promoter followed by the AttBl recombination sequences and then an AAA sequence that precedes the initiating ATG. The sequence of the first two codons was also modified (sequence GAA-CCT after the ATG initiation codon).

The final plasmid carrying the DHCR24 cDNA under the control of the CYC1 promoter and also the S. cerevisiae 2u origin of replication and the URA3-d selection marker was named pIM331. Its equivalent without the DHCR24 cDNA was cloned into pIM303.

The YIM59 strain is transformed independently with plasmids pIM303 and pIM331, and two transformants carrying plasmid pIM303 (YIM59/pIM303 strain) or pIM331 (YTM59/pIM331 strain) are more specifically selected.

These strains are grown in a reconstituted Kappeli-type rich medium for 72 hours at 28°C to achieve an absorbance of 40 at 600 nm. Total sterol extracts (esterified sterols and free sterols) of the YIM59/pIM331 strain (which does not carry the DHCR24 cDNA) and of the YIM59/pIM331 strain (which carries the DHCR24 cDNA) are produced in the presence of methanolic potassium hydroxide (cf. Examples 5 and 6). These two strains are tested for their ability to produce cholesterol. Some of the results (GC) are given in Figure 13. The retention times presented are given in minutes on two chromatograms.

It was thus possible to show that the strain which does not carry the vector to express DHCR24 (the YIM59/pIM303 strain) does not produce cholesterol (part A), but mainly desmosterol (part A). In contrast, the strain carrying the DHCR24 cDNA (the YIM59/pIM331 strain) produces a sterol that has the retention time of cholesterol (Part B). It was possible to show, using techniques such as gas chromatography coupled to electron impact mass spectrometry (as described in Example 6), that this sterol is indeed cholesterol. By using the surface areas of each of the sterol peaks, it was possible to estimate that the amount of cholesterol produced by the YIM59/pIM331 strain was 57% of the sterols.

Disposal of biological material

The following organisms were deposited, on 22 April 2004, at the Collection Nationale de Cultures de Microorganismes (CNCM) [National Collection of Microorganism Cultures], 25 rue du Docteur Roux, 75724 Paris Cedex 15, France, in accordance with the provisions of the Treaty of Budapest.

- The WGIF04 strain deposited under accession number 1-3203.

All the publications and patents mentioned are incorporated into the present application by reference.

Bibliography:

- Adams, A.K., and C. Holm. 1996. Mol CellBiol. 16:4614-20.

- Ausubel Fred M., Brent Roger, Kingston Robert E., Moore David D., Seidman J.G., Smith John A.,

Struhl Kevin (Editors). Current Protocols in Molecular Biology, published by John Wiley & Sons Inc. Current Protocols Customer Service, 605 Third Avenue, 9th Floor New York, NY 10158 USA

(publication updated in March 2002).

- Baudin-Baillieu, A., E. Guillemet, C. Cullin, and F. Lacroute. 1997. Yeast. 13:353-6.

- Bonneaud, N, O. Ozier-Kalogeropoulos, G.Y. Li, M. Labouesse, L. Minvielle-Sebastia, and F. Lacroute. 1991. Yeast. 7:609-15.

- Bowman, S. et al. 1997. Nature. 387:90-3.

- Duport, C, B. Schoepp, E. Chatelain, R. Spagnoli, B. Dumas, and D. Pompon. 2003. Eur JBiochem. 270:1502-14.

- Gietz, R.D., RH. Schiestl, A R. Willems, and RA Woods. 1995. Yeast. 11:355-60.

- Goffeau, A., et al. 1996. Science. 274:546, 563-7.

- Jensen-Pergakes, K.L., M.A. Kennedy, N.D. Lees, R. Barbuch, C. Koegel, and M. Bard. 1998.

Antimicrob Agents Chemother. 42:1160-7.

- Kaneshiro, E.S., J.A. Rosenfeld, M. Basselin-Eiweida, J.R. Stringer, S.P. Keely, A.G. Smulian, and J.L.

Gins. 2002. Mol Microbiol 44:989-99.

- Kappeli, O., M. Arreguin, and M. Rieger. 1985. J Gen Microbiol. 131:1411-6.

- Kelly, S.L., D.C. Lamb, B.C. Baldwin, A.J. Corran, and D.E. Kelly. 1997. JBiol Chem. 272:9986-8.

- Lecain, E., X. Chenivesse, R. Spagnoli, and D. Pompon. 1996. JBiol Chem. 271:10866-73.

- McCammon, M.T., M.A. Hartmann, CD. Bottema, and L.W. Parks. 1984. JBacteriol. 157:475-83.

- Moebius, F.F., B.U. Fitzky, J.N. Lee, Y.K. Paik, and H. Glossmann. 1998. Proe NatlAcad Sei U SA. 95:1899-902.

- Noda, H., and Y. Koizumi. 2003. Insect Biochem Mol Biol. 33:649-58.

- Ozier-Kalogeropoulos, O., A. Malpertuy, J. Boyer, F. Tekaia, and B. Dujon. 1998. Nucleic Acids Res. 26:5511-24.

Skaggs, B.A., J.F. Alexander, CA. Pierson, K.S. Schweitzer, K.T. Chun, C. Koegel, R. Barbuch, and M.

Bard. 1996. Genes. 169:105-9.

- Sturley, S.L. 2000. Biochim Biophys Acta. 1529:155-63.

- Szczebara, F.M., et al. 2003. NatBiotechnol. 21:143-9.

- Tait, E., M.C. Simon, S. King, A.J. Brown, NA. Gow, and D.J. Shaw. 1997. Fungal Genet Biol. 21:308-14.

- Taton, M., and A. Rahier. 1991. Biochem Biophys Res Commun. 181:465-73.

- Taylor, F.R., and L.W. Parks. 1978. JBacteriol. 136:531-7.

- Waterham, H.R, J. Koster, G.J. Romeijn, RC Hennekam, P. Vreken, H.C Andersson, D.R. FitzPatrick,

RIDE. Kelley, and RJ. Wanders. 2001. Am JHum Genet. 69:685-94.

- Wood, V., et al. 2002. Nature. 415:871-80.

Claims (3)

1. Yeast strain that produces desmosterol, characterized in that it does not have a functional copy of the gene that codes for the enzyme 24-C-methyltransferase (ergå) and in that it expresses the heterologous enzyme 7-dehydrocholesterol reductase. 2. Yeast strain according to claim 1, characterized in that the gene encoding the enzyme 24-C-methyltransferase is inactivated. 3. The use of a yeast strain according to any one of claims 1 or 2 for the production of desmosterol.